<div dir="ltr"><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small;color:#000000">As indicated in my original reply, my interest in this project grows from my relative ignorance of Deep Learning. My career has focussed exclusively on symbolic computing. I've worked with and taught (a) functional programming, logic programming, and related issues in advanced Python; (b) complex systems, agent-based modeling, genetic algorithms, and related evolutionary processes, (c) a bit of constraint programming, especially in MiniZinc, and (d) reinforcement learning as Q-learning, which is reinforcement learning without neural nets. I've always avoided neural nets--and more generally numerical programming of any sort. </div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small;color:#000000"><br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small;color:#000000">Deep learning has produced so many impressive results that I've decided to devote much of my retirement life to learning about it. I retired at the end of Spring 2022 and (after a break) am now devoting much of my time to learning more about Deep Neural Nets. So far, I've dipped my brain into it at various points. I think I've learned a fair amount. For example, </div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small;color:#000000"><ul><li>I now know how to build a neural net (NN) that adds two numbers using a single layer with a single neuron. It's really quite simple and is, I think, a beautiful example of how NNs work. If I were to teach an intro to NNs I'd start with this.</li><li>I've gone through the Kaggle Deep Learning sequence mentioned earlier. </li><li>I found a paper that shows how you can approximate any differentiable function to any degree of accuracy with a single-layer NN. (This is a very nice result, although I believe it's not used explicitly in building serious Deep NN systems.)</li><li>From what I've seen so far, most serious DNNs are built using Keras rather than PyTorch.</li><li>I've looked at Jeremy Howard's <a href="http://fast.ai">fast.ai</a> material. I was going to go through the course but stopped when I found that it uses PyTorch. Also, it seems to be built on <a href="http://fast.ai">fast.ai</a> libraries that do a lot of the work for you without explanation. And it seems to focus almost exclusively on Convolutional NNs. </li><li>My impression of DNNs is that to a great extent they are <i>ad hoc</i>. There is no good way to determine the best architecture to use for a given problem. By architecture, I mean the number of layers, the number of neurons in each layer, the types of layers, the activation functions to use, etc. </li><li>All DNNs that I've seen use Python as code glue rather than R or some other language. I like Python--so I'm pleased with that.</li><li>To build serious NNs one should learn the Python libraries Numpy (array manipulation) and Pandas (data processing). Numpy especially seems to be used for virtually all DNNs that I've seen. </li><li>Keras and probably PyTorch include a number of special-purpose neurons and layers that can be included in one's DNN. These include: a DropOut layer, LSTM (short-long-term memory) neurons, convolutional layers, recurrent neural net layers (RNN), and more recently transformers, which get credit for ChatGPT and related programs. My impression is that these special-purpose layers are <i>ad hoc</i> in the same sense that functions or libraries that one finds useful in a programming language are <i>ad hoc</i>. They have been very important for the success of DNNs, but they came into existence because people invented them in the same way that people invented useful functions and libraries. </li><li>NN libraries also include a menagerie of activation functions. An activation function acts as the final control on the output of a layer. Different activation functions are used for different purposes. To be successful in building a DNN, one must understand what those activation functions do for you and which ones to use. </li><li>I'm especially interested in DNNs that use reinforcement learning. That's because the first DNN work that impressed me was DeepMind's DNNs that learned to play Atari games--and then Go, etc. An important advantage of Reinforcement Learning (RL) is that it doesn't depend on mountains of labeled data. </li><li>I find RL systems more interesting than image recognition systems. One of the striking features of many image recognition systems is that they can be thrown off by changing a small number of pixels in an image. The changed image would look
to a human observer just like the original, but it might fool a trained NN into labeling the image as a banana rather than, say, an automobile, which is what it really is. To address this problem people have developed Generative Adversarial Networks (GANs) which attempt to find such weaknesses in a neural net during training and then to train the NN not to have those weaknesses. This is a fascinating result, but as far as I can tell, it mainly shows how fragile some NNs are and doesn't add much conceptual depth to one's understanding of how NNs work. </li></ul><div>I'm impressed with this list of things I sort of know. If you had asked me before I started writing this email I wouldn't have thought I had learned as much as I have. Even so, I feel like I don't understand much of it beyond a superficial level. </div><div><br></div><div class="gmail_default"><div>So far I've done all my exploration using Google's Colab (Google's Python notebook implementation) and Kaggle's similar Python notebook implementation. (I prefer Colab to Kaggle.) Using either one, it's super nice not to have to download and install anything!</div><div><br></div></div><div style="color:rgb(34,34,34);font-family:Arial,Helvetica,sans-serif"><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div><div>I'm continuing my journey to learn more about DNNs. I'd be happy to have company and to help develop materials to teach about DNNs. (Developing teaching materials always helps me learn the subject being covered.)</div><div><br></div></div><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><font><u style="color:rgb(33,33,33);font-family:"Helvetica Neue",Helvetica,Arial,sans-serif;font-size:16.5px;line-height:20px"> </u></font><span style="color:rgb(33,33,33);font-family:"Helvetica Neue",Helvetica,Arial,sans-serif;font-size:16.5px;line-height:24.75px"> </span>-- Russ Abbott <br>Professor Emeritus, Computer Science<br>California State University, Los Angeles<br></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Jan 8, 2023 at 1:48 AM glen <<a href="mailto:gepropella@gmail.com">gepropella@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Yes, the money/expertise bar is still pretty high. But TANSTAAFL still applies. And the overwhelming evidence is coming in that specific models do better than those trained up on diverse data sets, "better" meaning less prone to subtle bullsh¡t. What I find fascinating is tools like OpenAI *facilitate* trespassing. We have a wonderful bloom of non-experts claiming they understand things like "deep learning". But do they? An old internet meme is brought to mind: "Do you even Linear Algebra, bro?" >8^D<br>
<br>
On 1/8/23 01:06, Jochen Fromm wrote:<br>
> I have finished a number of Coursera courses recently, including "Deep Learning & Neural Networks with Keras" which was ok but not great. The problems with deep learning are<br>
> <br>
> * to achieve impressive results like chatGPT from OpenAi or LaMDA from Goggle you need to spend millions on hardware<br>
> * only big organisations can afford to create such expensive models<br>
> * the resulting network is s black box and it is unclear why it works the way it does<br>
> <br>
> In the end it is just the same old back propagation that has been known for decades, just on more computers and trained on more data. Peter Norvig calls it "The unreasonable effectiveness of data"<br>
> <a href="https://research.google.com/pubs/archive/35179.pdf" rel="noreferrer" target="_blank">https://research.google.com/pubs/archive/35179.pdf</a><br>
> <br>
> -J.<br>
> <br>
> <br>
> -------- Original message --------<br>
> From: Russ Abbott <<a href="mailto:russ.abbott@gmail.com" target="_blank">russ.abbott@gmail.com</a>><br>
> Date: 1/8/23 12:20 AM (GMT+01:00)<br>
> To: The Friday Morning Applied Complexity Coffee Group <<a href="mailto:friam@redfish.com" target="_blank">friam@redfish.com</a>><br>
> Subject: Re: [FRIAM] Deep learning training material<br>
> <br>
> Hi Pieter,<br>
> <br>
> A few comments.<br>
> <br>
> * Much of the actual deep learning material looks like it came from the Kaggle "Deep Learning <<a href="https://www.kaggle.com/learn/intro-to-deep-learning" rel="noreferrer" target="_blank">https://www.kaggle.com/learn/intro-to-deep-learning</a>>" sequence.<br>
> * In my opinion, R is an ugly and /ad hoc/ language. I'd stick to Python.<br>
> * More importantly, I would put the How-to-use-Python stuff into a preliminary class. Assume your audience knows how to use Python and focus on Deep Learning. Given that, there is only a minimal amount of information about Deep Learning in the write-up. If I were to attend the workshop and thought I would be learning about Deep Learning, I would be disappointed--at least with what's covered in the write-up.<br>
> <br>
> I say this because I've been looking for a good intro to Deep Learning. Even though I taught Computer Science for many years, and am now retired, I avoided Deep Learning because it was so non-symbolic. My focus has always been on symbolic computing. But Deep Learning has produced so many extraordinarily impressive results, I decided I should learn more about it. I haven't found any really good material. If you are interested, I'd be more than happy to work with you on developing some introductory Deep Learning material. <br>
> <br>
> -- Russ Abbott<br>
> Professor Emeritus, Computer Science<br>
> California State University, Los Angeles<br>
> <br>
> <br>
> On Thu, Jan 5, 2023 at 11:31 AM Pieter Steenekamp <<a href="mailto:pieters@randcontrols.co.za" target="_blank">pieters@randcontrols.co.za</a> <mailto:<a href="mailto:pieters@randcontrols.co.za" target="_blank">pieters@randcontrols.co.za</a>>> wrote:<br>
> <br>
> Thanks to the kind support of OpenAI's chatGPT, I am in the process of gathering materials for a comprehensive and hands-on deep learning workshop. Although it is still a work in progress, I welcome any interested parties to take a look and provide their valuable input. Thank you!<br>
> <br>
> You can get it from:<br>
> <a href="https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0" rel="noreferrer" target="_blank">https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0</a> <<a href="https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0" rel="noreferrer" target="_blank">https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0</a>><br>
> <br>
<br>
-- <br>
\ua66e M\u0265\u01ddu \u01ddl\u01ddd\u0265\u0250u\u0287s \u025f\u1d09\u0183\u0265\u0287' \u0287\u0265\u01dd \u0183\u0279\u0250ss sn\u025f\u025f\u01dd\u0279s\u02d9 \ua66e<br>
<br>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .<br>
FRIAM Applied Complexity Group listserv<br>
Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer" target="_blank">https://bit.ly/virtualfriam</a><br>
to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
archives: 5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a><br>
1/2003 thru 6/2021 <a href="http://friam.383.s1.nabble.com/" rel="noreferrer" target="_blank">http://friam.383.s1.nabble.com/</a><br>
</blockquote></div>